Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Integrating relevance feedback techniques for image retrieval using reinforcement learning.

Peng-Yeng Yin1, Bir Bhanu, Kuang-Cheng Chang

  • 1Department of Information Management, National Chi Nan University, 303 University Rd., Puli, Nantou 545, Taiwan. pyyin@ncnu.edu.tw

IEEE Transactions on Pattern Analysis and Machine Intelligence
|October 22, 2005
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Toward Generative Understanding: Incremental Few-Shot Semantic Segmentation With Diffusion Models.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026
Same author

Sampling Conformational Ensembles of Highly Dynamic Proteins via Generative Deep Learning.

Journal of chemical information and modeling·2025
Same author

Computational modeling of tumor invasion from limited and diverse data in Glioblastoma.

Computerized medical imaging and graphics : the official journal of the Computerized Medical Imaging Society·2024
Same author

Sampling Conformational Ensembles of Highly Dynamic Proteins via Generative Deep Learning.

Research square·2024
Same author

Sampling Conformational Ensembles of Highly Dynamic Proteins via Generative Deep Learning.

bioRxiv : the preprint server for biology·2024
Same author

RepSGG: Novel Representations of Entities and Relationships for Scene Graph Generation.

IEEE transactions on pattern analysis and machine intelligence·2024
Same journal

Relation DETR+: Exploring Explicit Position Relation Prior for Dense Prediction.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

RBF++: Quantifying and Optimizing Reasoning Boundaries across Measurable and Unmeasurable Capabilities for Chain-of-Thought Reasoning.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

CAFE: Cross-View Adaptive Fusion and Cluster Center Enhancement for Robust Multi-View Clustering.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

DIVER: Reinforced Diffusion Breaks Imitation Bottlenecks in End-to-End Autonomous Driving.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

Ethics-Aware Safe Reinforcement Learning for Rare-Event Risk Control in Interactive Urban Driving.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

Learning Shape Anchors for Holistic Indoor Scene Understanding.

IEEE transactions on pattern analysis and machine intelligence·2026
See all related articles

This study introduces an image relevance reinforcement learning (IRRL) model to integrate existing relevance feedback (RF) techniques for better content-based image retrieval. Integrating multiple RF methods and sharing knowledge improves performance and reduces storage demands.

Area of Science:

  • Computer Science
  • Information Retrieval
  • Machine Learning

Background:

  • Relevance feedback (RF) is crucial for refining search results in information retrieval systems.
  • Existing research often focuses on novel RF techniques, overlooking the synergy of integrating established methods.
  • Content-based image retrieval (CBIR) systems benefit significantly from effective relevance feedback mechanisms.

Purpose of the Study:

  • To propose an image relevance reinforcement learning (IRRL) model for integrating diverse RF techniques within a CBIR system.
  • To enhance retrieval performance by leveraging multiple RF approaches and shared user experiences.
  • To address storage complexity challenges in large-scale image retrieval databases.

Main Methods:

  • Development of an image relevance reinforcement learning (IRRL) model.

Related Experiment Videos

  • Implementation of various integration schemes for combining existing RF techniques.
  • Utilizing a long-term shared memory for multi-user retrieval experience exploitation.
  • Introduction of a concept digesting method for reducing storage complexity.

    • Main Results:

    • The integration of multiple RF approaches yielded superior retrieval performance compared to single-technique usage.
    • Sharing relevance knowledge across multiple query sessions significantly boosted retrieval accuracy.
    • The concept digesting technique effectively reduced storage demands, demonstrating model scalability.
    • The IRRL model proved more effective than traditional RF methods.

    Conclusions:

    • Integrating multiple relevance feedback techniques through an IRRL model enhances CBIR performance.
    • Shared relevance knowledge and efficient storage methods are key to scalable image retrieval.
    • The proposed IRRL model offers a robust and efficient solution for advanced content-based image retrieval.